Log in

No account? Create an account

Previous Entry | Next Entry

HCoFP: finis

No, I haven't finished the book yet. Working on that right now. Final chapters were a recap of everything else, skipping. Some appendices of or more detailed calculations. Then, appendix F! He excoriates scientific urban legend [my words], wherein a factoid is cited by a whole chain of people, with no clear source. In this case it was "US cities devote half their space to cars; in Los Angeles, two-thirds". Even Shoup's sceptical of this. Actual data?

[Me thinking: Well, given that building codes typically require as much or more parking space as building floor area, it's hardly implausible that you've got as much land in parking as building, then you add parking to the streets...]

Well, *streets* are funny. For one thing, they pre-date cars (at least in older cities) and can't be credited entirely to them, though of course they took over streets through the invention of jaywalking. More concretely, the percentage of land area spent as streets is 30 in New York, 26 in San Francisco, 24 in Chicago, but only 13 to 15 in Dallas, Houston, Atlanta, or Los Angeles.

Surprising? Not so much if you think about the layout. Big long blocks and deep lots, vs. lots of short blocks; pedestrian friendliness requires a lot of cuts (though they needn't be wide ones). A more concrete statistic is street area per capita (1983 numbers), which ranges from 257 square feet in Newark (fairly dense at 17,000/mile2, with only 16% in streets) to 345 NYC, 441 San Francisco, 741 LA, or 1575 Dallas. So car cities spend about half as much urban space on cars, but this is being used by as few as 1/5 as many people.

Of course, that's just streets. What about parking? Remembering that the average car spends 95% of its time parked, and that people don't all drive at the same time, we can see that we might need more space to parking than to driving. A car parked in a lot needs 30 square meters, a car at 20 meter/s (44 mph) and two second following distance might need 86 m2 [1] but the car needs the road space for only an hour two a day.

But then he goes and talks about urban density, like the fact that the Census Bureau urbanized area of LA is denser than that of NY, since NY looks like "Hong Kong surrounded by Phoenix" while LA's area looks like a lot more LA. LA suburb density is 82% of core density, for NY that's 12%. He suggests that LA is about as dense as a city can get while being devoted to cars -- I'd wonder how much of the underclass actually takes buses or carpools -- and notes the 14% of land to streets, though with population growth that's only 483 square feet per capita in 2000. "No one knows how much land is devoted to parking." Welp.

[1] width 1.7 m, time swept area of 40 m (20*2), plus the area of the car (8-11 m2, I used 9).

Appendix G is on congestion, and there's a table of LA data. It's pretty non-linear; at low densities like 10 cars per mile, tripling the density has little effect. Well, actually kind of big: at 10 cars, you've got speed of 73 mph and 0.8 minutes/mile; at 30 you've got 53 mph and 1.1 minutes/mile. 20%, but not bad for tripling. But when you've got 130 cars per square miles, they take 5.9 minutes per mile, and going to 140 cars means they take 6.9 minutes per mile. Optimal traffic flow is about 1800 per hour in a lane (which allows for 2 second following distance); raising tolls as traffic approaches that point could keep roads moving quickly, "converting wasted time into real money."

As before, he suggest revenue sharing to overcome political obstacles, and also be fairer, with toll money going to cities that freeways pass through. (Fairness is shown from the fact that no one wants a freeway right near them, and they usually end up routed through poor cities or neighborhoods. The revenue also makes up for land that isn't generating property or sales tax. In LA County, the 66 cities with freeways had a per capita income of $20,000, while those without had an income of $35,000.) One estimate was that congestion tolls could raise $7 billion for LA in 2010; a different estimate was that the social costs of congestion were $12 billion in 2001! Even $5 billion would double the municipal revenue of the freeway cities. Higher-income residents drive more, including at peak hours, so it's "tax the rich" while helping them drive faster. If $1.1 billion were given to the MTA from tolls, the county sales tax could be reduced from 8.25% to 7.25%

Appendix H: in 2000, the US had 771 cars per 1000, people, the world had 121, the non-US world had 89. US had 30% of the world's vehicles, which is down a lot, from 75% in 1946.

The Afterword describes some real world parking reforms. First one is DC, which in 2008 changed meter policies near a ballpark of 41,000 seats and 1300 off-street parking spaces. Aiming for 10-20% vacancy, the charge $8/hour during ballpark events and $2/hour the rest of the day; $1-1.50/hour on non-game days. 75% of revenue goes back to the neighborhoods.

NYC in 2008 tried PARK Smart in Greenwich Village. Rates jumped from $1/hour (!!!!) all day to in $2/hour between noon and 4; 61% of drivers and 57% of merchants said parking remained easier or about the same. Only 46 and 34% were even aware of the new rates. Brookelyn 2009: rates went from 75 cents (!!!!!!) to $1.50 between nnon and 4; peark parking durations went down 20%, number of parked vehciles went up 18%, traffic volume went down by 7% (less cruising.)

Seattle's also trying; they got merchant support by switching from revenue goals to outcome goals (1-2 vacant spaces per block.)

Chicago completely bungled its privatization from a public policy perspective, but even with limits on rate increases, the winning bid was $1.16 billion for 36,000 metered spaces on a 75 year concession. So spaces are worth at least $32,000 apiece.

Enforcement: a small percentage of drivers generate large shares of parking tickets (8 and 29%, 5 and 24%, 5 and 22%, 14 and 47%); he suggests graduated fines. For first offense, a warning, for a few more, modest fines, but steep after that, to discourage people from treating fines as just the price they pay. In Claremont, CA overtime parking fines are $435, 70, and $105 for 1st-3rd offenses; handicapped fines are $325, 650, $975. Citations fell 22% in the year after such fines were introduced, with repeat citations falling the most (31% for cars with 4 or more violations.) Wireless handheld devices help officers write the right ticket. Warning on first offense is particularly appropriate if you're changing the rules on people.

NYC is trying self-release boots on cars with $350 in tickets; if you pay via phone, the boot unlocks, though you still have to return it to the city.

Arlington VA got rid of disability placard exemptions from parking meters, but reserves some end spaces for disabilities. Idea is to keep convenient spaces open for them, not to let them park for free.

Alexandra V found that 90% of handicapped placards and plates were being used illegally.

Interesting idea: if you're going to have parking lots, requiring them to have solar power canopies works well with them. Has public justification -- more parking spaces implies more customers means more power demand -- and it's easier to get panels facing the right way on an open lot than on the roof of a house.

Boulder unbundles, requiring residential parking to be leased separately from the housing. Bellevue does that somewhat, and also requires the monthly rate to not be less than the cost of a transit pass.

Finally done.

See the comment count unavailable DW comments at http://mindstalk.dreamwidth.org/391555.html#comments



Damien Sullivan

Latest Month

September 2017


Powered by LiveJournal.com
Designed by Lilia Ahner